TRAVEL COST METHOD (TCM)

Explain the concept of travel cost

method

Prepare questionnaire to be used in

TCM

Apply TCM to calculate recreation

benefits

Learning Outcomes

Introduction

TCM is used to value recreational uses of the environment

It is commonly applied in benefit cost analyses (BCA) and in natural resource damage assessments

It is based on „observed behaviour‟, thus is used to estimate use values only

TCM is a demand-based model for use of a recreation site or sites

A site: a river for fishing, a trail for hiking, a forest camping/recreation, a park for wildlife viewing, a beach for swimming, a marine park for snorkeling, an urban park for recreation, etc

Criteria For Recreation Valuation

Procedures

For the efficient allocation of resources, procedure for estimating

contribution of recreation should meet the following criteria:

Estimates of value – consistent with and have a level of precision similar to the estimates of value derived for other goods and services produced by alternative plans

Procedures – applicable to evaluate proposed changes in the availability of the specific recreation opportunities affected by the projects being analysed

Estimates of values of existing sites useful if the analysis is used to develop models to value a proposed change in the availability of a similar opportunities

Individuals facing an easily accessible range of highly desirable alternatives will presumably be willing to pay less for use of a particular area than individuals with fewer and less desirable alternatives

Criteria For Recreation

Valuation Procedures…..cont’d

Different values may be placed on recreation participation by different subsets of total population (may be based on income, past participation, family structure, distance, and other factors).

If significant, they should be included in the estimation procedure

The total level of use may be due to crowding and congestion and this affect the value of the site to an individual.

Changes in the value due to different levels of congestion are appropriately attributed to the alternative that induced them.

Economic Theory

Useful to separate between single site model and multiple site model

For a single site model, it works like a conventional downward sloping demand functions

The „quantity demanded‟ – the number of trips taken to a recreation site

The „price‟ – the trip cost of reaching the site

Variation in price is generated by observing people living in different distances from the site

Price is low for people near the site and high for those living farther away

Therefore, the demand function slopes downward if trips decline with a distance to the site. Refer to diagram

Demand Curve and Consume Surplus

Price

(RM/unit)

QuantityO

Consumer surplus when price actually

paid at P1 (Area A)

If price paid is P2, consumer surplus is

Area A + B + C

If price paid is zero, consumer surplus

is area under demand curve

V2

P1

V1

P2

Q2Q1

B

A

C Demand curve:

Q = f(P)

Demand curve for recreation

Vi is the total number of

trips by individuals i to

the site in a period of

time

TCi is the trip cost to the

site

Trip cost

(RM/visit)

O V2V1

Trips

TC

V=f(TC)

V3

TC1

TC2

TC3

Vi = f(TCi)

Consumer surplus

(Area Under the curve) a

bdTCbTCaCS )(

Area under the curve = Consumer surplus

(a measure of social welfare)

Choke

price

3

TCM

Travel cost method (TCM) is one of the popular methods used falls under the revealed preference approach to value non-priced or non-market good

The other two approaches are: market-based and stated preference approach

This approach is based on the assumption that the non-marketed good or service affect preferences expressed by consumers

Preference of consumers are expressed in terms of expenditure of marketed good and service (travel expenditure), or in the level of productivity by certain market activities (wage rate or house price)

The purpose is to identify the relationship between environmental change and the affected prices of marketed goods and services

Basic Concepts

3

TCM

How this is done?

Through applications of some techniques based on household production function

Consumers are willing to pay for a good or service because they want to maximize utility (satisfaction)

This utility is reflected in the expenditure that the consumers spend in obtaining this good or service

Basic Concepts…..

3

TCM

Basic Concepts

By relating the number of visits per period to travel costs, a demand curve can be estimated that can be used to estimate the WTP of individuals for recreational services of the whole site

The consumer surplus is used as a proxy of the WTP

Survey of visitors is carried out using a structured questionnaire

3

TCM A sample of visitors to a site is interviewed to ascertain the characteristics of their visits to the site – Frequency of visits per period Travel cost to the site Origin of visitors Socio-economic information (age, income, education, gender,

household size, etc) Purpose of visits

The travel costs (including time cost) is a proxy for market prices in demand estimation

Two variants of travel cost method: Single site model (Individual Travel Cost Method (ITCM) and Zonal Travel

Cost Method (ZTCM). Individual travel cost model is the most popular method.

Multiple site model (Random Utility Maximization (RUM) Model or survey method – discrete choice)

Basic Concepts…..

Taxonomy of Travel Cost Method

Approach of

Economic Valuation

Market-based Revealed Preference Stated Preference

Travel Cost Method Hedonic Price MethodWage Differential

Approach

Single Site Model Multiple Site Model

Individual Travel Cost Method (ITCM)

Zonal Travel Cost Method (ZTCM)

Random Utility Maximization Model (RUM)

(also known as survey method)

Single Equation

Systems of Equation

(Hedonic Travel Cost Method)

CONSUMER THEORY OF TCM

Max: u ( x, v, q ) (1)

s.t. Monetary and Time Constraints:

m + pw.tw = x + c . v (2)

t* = tw + (t1 + t2) v (3)

Where

x = quantity of other goods consumed

v = number of trips to the site

q = environmental quality at the site

m = exogenous income

pw = wage rate

c = monetary cost of a trip

t* = total discretionary time

tw = hours worked

t1 = round trip travel time and t2 = time spent on site

CONSUMER THEORY

Two components of monetary cost for the

trip: admission fee (f)

the monetary cost of distance travel (pd.d) where

- pd is the travel cost/km.

- d is the distance to and from the site.

Opportunity cost to time spent: wages

foregone

Substituting the time constraint (3) into

budget constraint (2) yields (4):

m + pw.t* = x + tc.v (4)

CONSUMER THEORY

Where tc is the full price of a visit given bytc = c + pw ( t1 + t2 )

= f + pd.d + pw (t1 + t2 ) (5)

Four components of tc: admission fee

monetary cost of travel

time cost of travel

time cost at site

Maximizing (1) s.t. Constraint (4) leads to an

indvidual‟s demand or trip generation function (TGF) for visit v=f(M,tc,q)

SINGLE SITE MODEL

Single Site Model Estimate total use of “access value”

Use to estimate Loss value – total surplus, i.e. difference between a person‟s

total WTP for trips and the actual trip cost over a season.

Examples: beach close due to oil spill, forest recreation close due to development, development project eliminate wildlife viewing

Value associated with a change in the cost of access to a site. E.g. increase in an entry fee, opening of new entrance, addition of new site (for example, reservoir created by a dam -more difficult to value this. Can use benefit transfer approach to estimate the value of this site from another site of similar characteristics).

Can also use to value changes in site characteristics (e.g. improved water quality on a lake, increase in the number hiking trails in a wilderness area). But this is the strength of single site model.

When the goal is to value changes in site characteristics or to vale access to more than one site simultaneously, multiple site model is preferred

Single Site Model

Site

Individual

Experience

(Satisfaction)

Individualdistance

distance

3

TCM

Is a demand model for trips to a recreation site by a person over a season

The ‘quantity demanded’ – the number of trips taken to a recreation site

The ‘price’ – the trip cost of reaching the site. Includes:Travel expenses

Access fees

Equipment cost

Time cost

Single Site Model- Introduction

3

TCM The simplest form of the model is:

V = f(tcv), v is number of trips taken by a person in a season to the site, and tcv is the trip cost of reaching the site

Relationship is negative

Demand will depend on other factors: income, age, experience in the recreation activities, proximity to other recreation sites

Thus, full model may be written as:

Vi = f(tcv, tcs, y, z)

tcs vector trip cost to other recreation sites, y is income, z is socio-economic variables of visitors i

tcs – positive coefficient, y – positive coefficient, z depends on variables included in the model

Linear equation: Vi = +1*tci+2*tcsi + 3* yi + 4*zi + i

Single Site Model- Basic Theory

Access value in a linear single-site model

Area A: total CS for trips to the site during the season: the difference between total willingness to pay for trips (Area A + B) d total trip cost (Area B)

This is called “individual‟s access value” for the site

If the site were closed for the season, individual would loose access to the site and consequently the area A

Area C: Deadweight loss (if some visitors are excluded from using the site)

Trip cost

(RM/visit)

O V2

Trips

TC

Vi = f(tcv, tcs, y, z)

TC

Vi = f(TCV,TCS, y, z)

Consumer surplus

(Area Under the curve)

a

bdTCbTCaCS )(

Choke

price

A

BC

Assumptions for demand

of a single site

Individual Perceives Changes in Travel-related Cost of Visit the Same Way As Changes to Admission Fee

Each Trip for Sole Purpose of Visiting Site (otherwise, problem of joint cost)

All visit, incurs the same amount of time at site. 2 roles:

- Allows measuring of site usage by the scalar „v‟.

- Full price of a visit is a parameter to the individual.

No utility or disutility derived from traveling time.

Wage rate is the relevant opportunity cost of time

No alternative recreational sites available.

If assumptions are not met, then the TCM model would

have to be respecified.

Typical Form of Data Set for

Estimating a Single Site Model

Observation No of trips to

the site for

the season

Trip Cost to

the site (RM)

Trip cost to

substitute

(RM)

Monthly

Income

(RM)

Years engaged

in this for of

recreation

(yrs)

No of

children

Age

(yrs)

1 7 45 200 1500 17 4 35

2 1 120 20 800 2 0 24

3 21 50 65 4000 22 5 40

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

N 3 98 65 1300 10 1 28

Steps Involved

Step Activity

1 Identification of sites of Recreation &

Ecotourism Interest

2 Questionnaire Design

3 Survey of Sampled Visitors

4 Database creation and data analysis

5 WTP Estimation (Estimate of Consumer

Surplus)

Steps Involved

1: Identification of sites

of Recreation &

Ecotourism Interest

1a: Assessment of

recreational potential of

the site

1b: Data collection of

visitors flow

2: Questionnaire Design

2a: Introduction

2b: Socio-economic

Information

2c: Questions on the

dependent variable and

explanatory variables

Steps

Steps Involved……3: Survey of Sampled

Visitors

3a: Sampling technique

3b: Survey strategy: How,

When, and Where to run

interviews

3c: Training of

Enumerators

3d: Pretest survey or pilot

survey

3e: Running the interviews

4: Database creation and

data analysis

4a: Verification of data

4b: Database creation

4c: Elimination of invalid

questionnaires and answers

(data cleaning)

4d: Derived variables

building

4e: Data analysis

Steps Involved……5: WTP Estimation (Access Value)

5a: Choice of ITCM equations (linear, log

linear, double log, exponential)

5b: Implement Models

5c: Annual individual Average WTP

5d: Annual net benefits

5e: Discounted value of annual benefits

SAMPLING INTENSITY

Representativeness of sampling frame covering

various

sub-populations of visitors

period of survey

This allows the use of „large number‟ statistic

tests such as t and F tests of significance.

Sampling intensity is dependent on sampling

error considered appropriate.

3

TCM

Trip cost: sum of the expenses required to make a trip

Typical costs for a day trip include: Travel cost (round trip cost which include all

transit expenses – fuel and upkeep, toll. Calculation: per-km cost*round-trip distance –RM/km)

Access fees (RM per person) Equipment cost (vary by type of recreation. Can

use rental fee. Often excluded because difficult to estimate)

Time cost (travel time and time spent on site, most difficult to estimate. Use wage rate –Calculation: wage rate/hr*time (hr). Use 1/3 of wage rate – lower bound. Upper bound use full wage rate)

Trip Cost

Estimate the Model: Some alternative and choice

of functional forms

Functional forms : Linear: Vi = a + b*TCi + c*TCSi + d*Yi+e*Zi

Semi-log: Ln V = a + b*TCi + c*TCSi + d*Yi+e*Zi

Double-log: Ln V = a+b*LnTCi+c*LnTCSi+d*LnYi+e*LnZi

Functional Selection A negative sign on the travel cost coefficient and

appropriate signs for other coefficients A statistically significant t and F statistic at the 95% level

of confidence An acceptable level of R2 statistic

Specify the Model

Before data collection begins, consider variables in the right hand side

V =f(TCv, TCS, Y, Z)

Every model include trip cost TC Other variables:

Trip cost to other sites Income Socio-economic variables

family size Age Gender urban/rural residence Occupation level of education club membership equipment ownership attitudinal information experience in activity

Some Empirical Issues

Choice of dependent variable

Multipurpose trips

Holiday-makers vs. Residents

Calculation of distance costs

The value of time

Statistical problems

Empirical Analysis

General model:

VISIT = f(TC,AGE,INCOME,EDUYR,GENDER,MARITAL)

Specific Model (Semi-log Model):

Estimated Model:

ii6i5i4i3i2i1i εMARITALβGENDERβEDUYRβINCOMEβAGEβTCβαLnVISIT

i6i5i4i3i2i1i MARITALβ̂GENDERβ̂EDUYRβ̂INCOMEβ̂AGEβ̂TCβ̂α̂ISITV̂Ln

Consumer surplus estimation

Consumer surplus is estimated from the area under the

demand curve

This is by integrating the demand equation with respect

to travel cost in the range between the choke travel price

and the mean travel price for the whole sample

The choke travel price is that cost above which there will

be no trip undertaken by visitors, i.e. upper truncation of

the travel cost axis

The value of the integration measures the area below

the demand curve in the range of the mean travel price

to the choke travel price

Computation of mean consumer’s surplus per visitor

Functional

form

Equation Mean consumer surplus/visitor

Linear

Semi-log

Double log

XTCV 21

XTCLnV 21

LnXLnTCLnV 21

*211 |

2)(| TC

TCTCTCX

*

1

|))((1

| 21 TC

TC

XTCee

*)1(

1

|)1

1(| 12 TC

TCTCX

Note:

TC* is the choke travel price, TC is the mean travel price and is the mean of the vector of socio-

economic variables

V is the number of visits per year

TC is travel price variable and X is the vector of socio-economic variables

, 1, 2 are the regression coefficients

e is the exponential sign taking the value of 2.718

X

MULTIPLE SITE MODEL

3

TCM

Introduction

RUM considers a person’s choice of a site for a

recreation trip

In choosing a site, a person is assumed to consider

its “price” and its characteristics

The “price” is the trip cost

The characteristics – attributes/features of the site:

ease of access, environmental quality, facilities

Time frame – “choice occasion”. E.g. when

analyzing day trips, a choice occasion is simply a

day

Multiple Site Model or Random Utility

Maximization (RUM) Model

3

TCM

Basic Theory

On a given choice occasion, a person considers

visiting one of the sites, C as i=1,2,3,…C

Each site – provide a person site utility Vi

Utility, Vi is a function of trip cost (tc) and site

characteristics (q)

Utility for site i, is given by:

Vi = + 1tci+2qi + i

Tc is trip cost to the site I, q is a vector of site

characteristics, is random error, is parameter

Multiple Site Model

Multiple Site Model

Site1 Origin Distance (km)

Characteristics

Site 3Site 2 Alternative

sites

Origin

Multiple Site Model

To value changes in site characteristics at one or more sites

To value the access to more than one site simultaneously

Use random utility maximization (RUM) model

The method is also known as survey method

Asking people about their choice of many recreation sites

Multiple Site Model……RUM

Individual I choice of one recreation site from

a set of many possible sites on a single

choice occasion in a season

The choice set depends on characteristics of

the area

Trade off one site for another trip cost

included as on characteristics

Multiple Site Model……RUM Normally use in quality improvement. Some examples,

include

Improvement in water quality on lakes/rives

Increase in catch rate of fish on lakes/rivers

Improvement in condition of access to several urban/recreation parks

Increase in wildlife population in several living areas

Increase in the number of mountain hiking trails in a state park

Improve access to one site or some sites simultaneously (e.g. to value the loss of several beaches closed because of development, to value several trails opened in the eco tourism development project)

Data Set for Estimating a RUM

Model (Three-site Choice Set)

Observ

ation

Number of Trips to Site Trip Cost To Site (RM) Water Quality Index at Site

(1-10)

Monthl

y

Income

Age

Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 (RM) (yrs)

(r1) (r2) (r3) (tc1) (tc2) (tc3) (q1) (q2) (q3)

1 2 0 17 45 158 15 10 2 1 1700 35

2 1 0 3 111 201 35 8 7 5 2500 40

3 0 3 0 29 33 345 2 8 9 1000 28

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

N 12 0 0 12 66 123 5 2 1 1700 32

Steps in Estimation - RUM

Step 1. Identify the impacts to be values

Step 2. Define the population of Users to be analyzed

Step 3. Define the choice set

Step 4. Develop a sampling strategy

Step 5. Specify the model

Step 6. Gather site characteristics

Step 7. Decide on the treatment of multiple purpose trips

Step 8. Design and implement the survey

Step 9. Measure trip cost

Step 10. Estimate model

Step 11. Calculate access and/or quality change values